Nonlinear predictive control based on NARMAX models

This paper introduces a new nonlinear predictive controller synthesis methodology based on NARMAX models. The control actions are computed based on multistep-head stochastic NARMAX predictors which depend explicitly on future control increments and are identified directly from experimental data. The proposed design methodology can deal effectively with measurement noise and load disturbances in a similar manner to that adopted in the classical GPC approach introduced by Clarke et al. Numerical simulations are given to demonstrate the effectiveness and robustness of the proposed approach.